Target scattering angle estimation method based on electromagnetic vortex modal information entropy

Abstract

The invention discloses a target scattering angle estimation method based on electromagnetic vortex modal information entropy. The method comprises the following steps: step 1, obtaining a complex form of a target vortex scattering field; 2, extracting the modal spectrum purity of the target vortex scattering field modal l according to the complex form of the target vortex scattering field; step 3, obtaining a target vortex scattering mode information entropy according to a mode spectrum purity calculation formula of the vortex mode l of the given circular ring region; and step 4, estimating a target scattering angle based on the target vortex scattering modal information entropy. According to the method, a new vortex dimension feature and an information entropy concept are utilized, a new target scattering echo feature quantity is introduced, the feature quantity represents a modal spectrum distribution rule, and the defect that a traditional signal is prone to being affected by complex electromagnetic environment noise intensity can be overcome to a certain extent; on this basis, a target scattering angle estimation model based on modal information entropy is constructed, high-precision estimation of the target scattering angle is achieved in combination with a neural network model, and the method has high practical value.

Description

Technical Field

[0001] The present invention belongs to the technical field of new-system radar target detection, and in particular relates to a target scattering angle estimation method based on electromagnetic vortex modal information entropy, which can be applied to target scattering angle estimation and other spatial eigenvalue estimation fields. Background Art

[0002] With the rapid development of electronic warfare, the military, civil aviation or other radar users have put forward higher requirements for radars, such as fast, efficient and high-precision estimation of space targets and tracking targets to meet real-time requirements. The target angle estimation problem plays an important role in space target tracking and identification. Traditional radar uses the amplitude and phase of the space target echo signal, which is easily affected by the complex electromagnetic environment noise intensity. The echo signal is submerged in the noise, making it difficult to achieve high-precision and effective estimation.

[0003] Vortex electromagnetic waves carry orbital angular momentum information and have a higher degree of freedom of information modulation than traditional electromagnetic waves, so they have great application potential in the field of radar target detection and imaging. As a cutting-edge technology, a large number of scholars have explored the orbital angular momentum characteristics of vortex electromagnetic wave radars: in the field of radar imaging, super-resolution imaging and high-precision imaging of high-speed targets can be achieved; in the field of wireless communications, communication capacity and frequency efficiency can be greatly increased; in the field of target detection, more abundant target rotation Doppler information can be perceived. In-depth understanding of the characteristics of new-system vortex electromagnetic wave radars can promote the development and application of imaging, perception, and communication.

[0004] In order to use vortex radar technology to estimate the target scattering angle, this patent introduces the concept of information entropy to describe the complex characteristics of the spatial distribution of the target vortex scattering modal spectrum. Informatics believes that information entropy is the probability of occurrence of a certain specific information. From the perspective of information dissemination, information entropy can represent the value of information, which refers to the uncertainty of information. The entropy corresponding to information with high information density is very low, and the entropy corresponding to information with low information density is higher. This patent innovatively equates different modal purities to the probability of occurrence of the mode, so that when the vortex modal information entropy value is large, it means that the spatial distribution of the modal spectrum information is obviously scattered, and the amount of information on the types of modal types contained is large; when the vortex modal information entropy value is small, it means that the spatial distribution of the modal spectrum information is relatively concentrated, and the specific modal value is relatively clear. At this time, the angle between the incident angle and the scattering angle is generally very small.

[0005] Traditional radars are based on the plane wave approximation condition in the target detection process. The echoes received by the radar are all the scattering of the plane wave by the target. The vortex electromagnetic wave has a spiral structured phase wavefront. The continuous change of the wavefront enables the vortex electromagnetic wave to distinguish the radar target in azimuth. Combining the concept of vortex modal information entropy, new eigenvalues ​​can be introduced for vortex radar imaging and detection applications.

[0006] In summary, it is necessary to study the target scattering angle estimation technology based on electromagnetic vortex modal information entropy to further improve the performance of vortex radar system. Summary of the invention

[0007] The technical problem solved by the present invention is: to make up for the defect that traditional signals are easily affected by the noise intensity of complex electromagnetic environments, and creatively propose a target scattering angle estimation technology based on electromagnetic vortex modal information entropy. This characteristic quantity characterizes the law of modal spectrum distribution, and combined with a deep neural network model, it can achieve high-precision estimation of the target scattering angle.

[0008] In order to achieve the above object, the present invention adopts the following technical solution:

[0009] A method for estimating target scattering angle based on electromagnetic vortex modal information entropy specifically comprises the following steps:

[0010] Step 1, obtaining the complex form of the target vortex scattering field;

[0011] Step 2, extracting the modal spectrum purity of the target vortex scattering field mode l according to the complex form of the target vortex scattering field obtained in step 1;

[0012] Step 3, according to the modal spectrum purity calculation formula of the vortex mode l of the given annular region obtained in step 2, obtain the target vortex scattering modal information entropy;

[0013] Step 4: Estimate the target scattering angle based on the target vortex scattering modal information entropy obtained in step 3.

[0014] Further, the specific operation of step 1 is: assuming that the wavelength of the vortex electromagnetic wave is λ, set the sampling step size λ2, and the number of sampling points in the X direction is N X , the number of points collected in the Y direction is N Y , forming N X ×N Y The target vortex scattering field space collection array, the lower left corner of the collection array is set as the origin (0,0), assuming that the position of any sampling point on the space collection array (X m ,Y n ) is the spatial amplitude of the vortex scattering at m,n , the phase is φ m,n , where X m =1,2,...N X , Y n =1,2,...N Y , then the target vortex scattering field is expressed in complex form as:

[0015]

[0016] Wherein, j is a complex number.

[0017] Further, the specific operation of step 2 is: according to the distribution of the target vortex scattering spatial amplitude collected in step 1, determine the sampling point number (X) of the minimum amplitude position of the central area of ​​the target vortex scattering field min ,Y min ), find the sampling point number (X min +d,Y min ), where d is the number of sampling points between the maximum and minimum values ​​of the vortex amplitude main ring position in the X direction, and the target vortex scattering field value E on the ring with radius d is extracted s (X Get ,Y Get ), the calculation formula of the modal spectrum purity of the vortex mode l in the annular area is:

[0018]

[0019] In the formula, C l is the modal spectral purity of the target vortex scattered field mode l, is the angular sampling interval of the ring, is the angle from the horizontal direction to any sampling point on the ring, (X Get ,Y Get ) is the coordinate position of any point on the ring, and the round(·) function rounds the value in the brackets to an integer.

[0020] Further, the specific operation of step 3 is: using the modal spectrum purity calculation formula obtained in step 2, calculate the vortex mode range [-L OAM ,L OAM ] Total 2L OAM +1 group of modal spectrum purity values, and calculate the sum of the purity values ​​of the modal spectra of all groups within the vortex mode range, and use the sum of the purity values ​​to normalize the purity value of the modal spectrum to obtain the equivalent occurrence probability of each modal component within the modal range. The formula is as follows:

[0021] C lp =C l / sum[C l ],lp=l∈[-L OAM ,L OAM ]

[0022] In the formula, lp represents the normalized mode of the corresponding mode l;

[0023] Then the information entropy of the target vortex scattering mode in the sampling ring area is:

[0024]

[0025] Further, the specific operation of step 4: the process of calculating the vortex scattering mode information entropy from step 1 to step 3 can be used to calculate any incident angle θ i , incident azimuth ψ i , scattering angle θ s , scattering azimuth ψ s The modal information entropy of the target vortex scattering modal information entropy database D is constructed. OAM ,Right now:

[0026] D OAM =[θ i ,ψ i ,θ s ,ψ s ,H OAM ]

[0027] In the formula, the minimum values ​​of the incident angle, incident azimuth, scattering angle and scattering azimuth are θ imin , imin ,θ smin , smin , and their maximum values ​​are respectively imax , imax ,θ smax , smax ,Right now:

[0028] θ i ∈[θ imin ,θ imax ],ψ i ∈[ψ imin ,ψ imax ],θ s ∈[θ smin ,θ smax ],ψ s ∈[ψ smin ,ψ smax ]

[0029] Using MATLAB deep neural network module, the database [θ i ,ψ i ,θ s ,ψ s ] as input, [θ s ] as the output, set the ratio of deep neural network training set, validation set and test set, as well as the number of hidden neural network layers according to the deep neural network module instruction, and perform training to obtain a trained deep neural network; input the [θ i ,ψ i ,ψ s ,H OAM ] parameter set, and the running result is the target scattering angle estimation.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] Aiming at the problem of vortex radar target scattering angle estimation, the present invention creatively proposes a target scattering angle estimation method based on electromagnetic vortex modal information entropy. Compared with the traditional radar target angle estimation method, the new dimensional characteristics of the vortex and the concept of information entropy are utilized to introduce a new characteristic quantity of the target scattered echo, which characterizes the law of modal spectrum distribution and can make up for the defect that the traditional signal is easily affected by the complex electromagnetic environment noise intensity to a certain extent. On this basis, a target scattering angle estimation model based on modal information entropy is constructed, and combined with a neural network model, high-precision estimation of the target scattering angle is achieved, which has high practical value. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is a block diagram of the target scattering angle estimation technology based on electromagnetic vortex modal information entropy;

[0033] Figure 2 The amplitude and phase spatial distributions of the vortex electromagnetic wave scattered field of mode 1 by the flat plate target, where (a) is the amplitude spatial distribution and (b) is the phase spatial distribution;

[0034] Figure 3 The modal spectrum distribution of vortex electromagnetic wave scattering mode 1 by the flat plate target;

[0035] Figure 4 The curve of the electromagnetic vortex modal information entropy changing with the incident angle and scattering angle;

[0036] Figure 5 The target scattering angle estimation result based on modal information entropy, where (a) is the fitting distribution of the prediction result and (b) is the error histogram. DETAILED DESCRIPTION

[0037] The implementation of the present invention is further described in detail below in conjunction with the accompanying drawings.

[0038] The application scenarios of the present invention are:

[0039] Aiming at the problem of vortex radar target scattering angle estimation, the present invention proposes a target scattering angle estimation technology based on electromagnetic vortex modal information entropy, and utilizes the new dimensional characteristic parameters of vortex modal spectrum to achieve high-precision target vortex scattering angle estimation, providing theoretical and technical reserves for vortex radar detection technology.

[0040] Step 1: Obtain the complex form of the target vortex scattering field.

[0041] Specifically, assuming that the wavelength of the vortex electromagnetic wave is λ, the sampling step is set to λ2, and the number of sampling points in the X direction is N X , the number of points collected in the Y direction is N Y , forming NX ×N Y The target vortex scattering field space collection array, the lower left corner of the collection array is set as the origin (0,0), assuming that the position of any sampling point on the space collection array (X m ,Y n ) is the spatial amplitude of the vortex scattering at m,n , the phase is φ m,n , where X m =1,2,...N X , Y n =1,2,...N Y , then the target vortex scattering field is expressed in complex form as:

[0042]

[0043] Wherein, j is a complex number.

[0044] Step 2: Extract the modal spectral purity of the target vortex scattering field mode l according to the complex form of the target vortex scattering field obtained in step 1.

[0045] Specifically, according to the distribution of the target vortex scattering spatial amplitude collected in step 1, the sampling point number (X) of the minimum amplitude position in the central area of ​​the target vortex scattering field is determined. min ,Y min ), find the sampling point number (X min +d,Y min ), where d is the number of sampling points between the maximum and minimum values ​​of the vortex amplitude main ring position in the X direction, and the target vortex scattering field value E on the ring with radius d is extracted s (X Get ,Y Get ), the calculation formula of the modal spectrum purity of the vortex mode l in the annular area is:

[0046]

[0047] In the formula, C l is the modal spectral purity of the target vortex scattered field mode l, is the angular sampling interval of the ring, is the angle from the horizontal direction to any sampling point on the ring, (X Get ,Y Get ) is the coordinate position of any point on the ring, and the round(·) function rounds the value in the brackets to an integer.

[0048] Step 3, according to the modal spectrum purity calculation formula of the vortex mode l of the given annular area obtained in step 2, obtain the target vortex scattering modal information entropy.

[0049] Specifically, the modal spectrum purity calculation formula obtained in step 2 is used to calculate the vortex modal range [-L OAM ,L OAM ] Total 2L OAM +1 group of modal spectrum purity values, and calculate the sum of the purity values ​​of the modal spectra of all groups within the vortex mode range, and use the sum of the purity values ​​to normalize the purity value of the modal spectrum to obtain the equivalent occurrence probability of each modal component within the modal range. The formula is as follows:

[0050] C lp =C l / sum[C l ],lp=l∈[-L OAM ,L OAM ] (5)

[0051] Where lp represents the normalized mode corresponding to mode l.

[0052] Then the information entropy of the target vortex scattering mode in the sampling ring area is:

[0053]

[0054] Step 4: Estimate the target scattering angle based on the target vortex scattering modal information entropy obtained in step 3.

[0055] Using the process of calculating the vortex scattering modal information entropy from step 1 to step 3, the arbitrary incident angle θ can be calculated i , incident azimuth ψ i , scattering angle θ s , scattering azimuth ψ s The modal information entropy of the target vortex scattering modal information entropy database D is constructed. OAM ,Right now:

[0056] D OAM =[θ i ,ψ i ,θ s ,ψ s ,H OAM ] (7)

[0057] In the formula, the minimum values ​​of the incident angle, incident azimuth, scattering angle and scattering azimuth are θ imin , imin ,θ smin , smin , and their maximum values ​​are respectively imax , imax ,θ smax , smax ,Right now:

[0058] θ i ∈[θimin ,θ imax ],ψ i ∈[ψ imin ,ψ imax ],θ s ∈[θ smin ,θ smax ],ψ s ∈[ψ smin ,ψ smax ] (8)

[0059] Using MATLAB deep neural network module (Neural Net fitting), the database [θ i ,ψ i ,ψ s ,H OAM ] as input, [θ s ] as output (Targets), set the ratio of deep neural network training set, validation set and test set, as well as the number of hidden neural network layers according to the deep neural network module instructions, and perform training to obtain the trained predictable scattering angle θ s The deep neural network of the trained deep neural network is input with the prediction [θ i ,ψ i ,ψ s ,H OAM ] parameter set, and the running result is the estimated value of the target scattering angle.

[0060] The effect of the present invention is further illustrated below by means of simulation data.

[0061] First, the process of calculating the electromagnetic vortex modal information entropy of the present invention is verified by simulation. When the vortex electromagnetic wave frequency is 16 GHz, the incident angle is 0 degrees, the incident azimuth angle and the scattering azimuth angle are both 0 degrees, and the scattering angle is 2 degrees, a high-frequency algorithm is used to simulate the spatial distribution of the scattering field of the vortex electromagnetic wave of mode 1 irradiating a flat target. Figure 2 (a) shows the spatial distribution of the target vortex scattering amplitude. Figure 2 (b) shows the spatial distribution of the target vortex scattering phase, and the horizontal and vertical coordinates are the size of the simulation sampling area. In Figure 2(a), the minimum amplitude is at the center position, and the minimum distance of the maximum value in the X direction is 0.84λ (the number of sampling point intervals is 13). According to the relationship between the sampling interval and the number of samples, the scattering mode spectrum distribution of the annular area with a radius of 0.84λ (the number of sampling point intervals is 13) is calculated as follows: Figure 3 As shown. Based on this, the incident angle is simulated from 0 to 66 degrees, the scattering angle is from 0 to 80 degrees, the angle interval is 2 degrees, the incident azimuth and the scattering azimuth are both 0 degrees, and the vortex mode information entropy database is constructed. There are a total of 1394 groups of data. Some data plot curves are shown in Figure 4As shown in the figure, the modal information entropy changes significantly under different scattering angles and incident angles, especially for the sampling surface parallel to the flat target, where the change with the incident angle is particularly significant. The MATLAB deep neural network module (NeuralNet fitting) was used. According to the module instructions, 70% of the data was used as a training set, 15% of the data was used as a validation set, and 15% of the data was used as a test set. The number of hidden neural network layers was set to 32. Figure 5 The results of scattering angle estimation using the modal information entropy database are given, with an R-squared determination coefficient of 0.9603 and an error of less than 2 degrees. The linear correlation and accuracy are both very high, and the target scattering angle can be estimated.

[0062] Conclusion of simulation analysis: For the new dimensional features of the new vortex radar mode, combined with the information entropy of the target vortex scattering mode extracted by the present invention, a deep neural network model is used to provide new ideas for the target vortex scattering angle estimation method.

Claims

1. A target scattering angle estimation method based on electromagnetic vortex modal information entropy, characterized in that: The specific steps include: Step 1, obtaining the complex form of the target vortex scattering field; Step 2, extracting the modal spectrum purity of the target vortex scattering field mode l according to the complex form of the target vortex scattering field obtained in step 1; Step 3, according to the modal spectrum purity calculation formula of the vortex mode l of the given annular region obtained in step 2, obtain the target vortex scattering modal information entropy; Step 4: Estimate the target scattering angle based on the target vortex scattering modal information entropy obtained in step 3.

2. The target scattering angle estimation method based on electromagnetic vortex modal information entropy as claimed in claim 1, characterized in that: Specific operation of step 1: Assume that the wavelength of the vortex electromagnetic wave is λ, set the sampling step size to λ2, and the number of sampling points in the X direction to N X , the number of points collected in the Y direction is N Y , forming N X ×N Y The target vortex scattering field space collection array, the lower left corner of the collection array is set as the origin (0,0), assuming that the position of any sampling point on the space collection array (X m ,Y n ) is the spatial amplitude of the vortex scattering at m,n , the phase is φ m,n , where X m =1,2,...N X , Y n =1,2,...N Y , then the target vortex scattering field is expressed in complex form as: Wherein, j is a complex number.

3. The target scattering angle estimation method based on electromagnetic vortex modal information entropy as claimed in claim 2, characterized in that: Specific operation of step 2: According to the distribution of the target vortex scattering spatial amplitude collected in step 1, determine the sampling point number (X) of the minimum amplitude position in the central area of ​​the target vortex scattering field min ,Y min ), find the sampling point number (X min +d,Y min ), where d is the number of sampling points between the maximum and minimum values ​​of the vortex amplitude main ring position in the X direction, and the target vortex scattering field value E on the ring with radius d is extracted s (X Get ,Y Get ), the calculation formula of the modal spectrum purity of the vortex mode l in the annular area is: In the formula, C l is the modal spectral purity of the target vortex scattered field mode l, is the angular sampling interval of the ring, is the angle from the horizontal direction to any sampling point on the ring, (X Get ,Y Get ) is the coordinate position of any point on the ring, and the round(·) function rounds the value in the brackets to an integer.

4. The target scattering angle estimation method based on electromagnetic vortex modal information entropy as claimed in claim 3, characterized in that: Specific operation of step 3: Use the modal spectrum purity calculation formula obtained in step 2 to calculate the vortex mode range [-L OAM ,L OAM ] Total 2L OAM +1 group of modal spectrum purity values, and calculate the sum of the purity values ​​of the modal spectra of all groups within the vortex mode range, and use the sum of the purity values ​​to normalize the purity value of the modal spectrum to obtain the equivalent occurrence probability of each modal component within the modal range. The formula is as follows: C lp =C l / sum[C l ],lp=l∈[-L OAM ,L OAM ] In the formula, lp represents the normalized mode of the corresponding mode l; Then the information entropy of the target vortex scattering mode in the sampling ring area is:

5. The target scattering angle estimation method based on electromagnetic vortex modal information entropy as claimed in claim 3, characterized in that: Specific operation of step 4: Use the process of calculating the vortex scattering modal information entropy from step 1 to step 3 to calculate any incident angle θ i , incident azimuth ψ i , scattering angle θ s , scattering azimuth ψ s The modal information entropy of the target vortex scattering modal information entropy database D is constructed. OAM ,Right now: D OAM =[θ i ,ψ i ,i s ,ψ s ,H OAM ] In the formula, the minimum values ​​of the incident angle, incident azimuth, scattering angle and scattering azimuth are θ imin , imin ,θ smin , smin , and their maximum values ​​are respectively imax , imax ,θ smax , smax ,Right now: i i ∈[θ imin ,i imax ],ψ i ∈[ψ imin ,ψ imax ],θ s ∈[θ smin ,i smax ],ψ s ∈[ψ smin ,ψ smax ] Using MATLAB deep neural network module, the database [θ i ,ψ i ,θ s ,ψ s ] as input, [θ s ] as the output, set the ratio of deep neural network training set, validation set and test set, as well as the number of hidden neural network layers according to the deep neural network module instruction, and perform training to obtain a trained deep neural network; input the [θ i ,ψ i ,ψ s ,H OAM ] parameter set, and the running result is the target scattering angle estimation.

Images